Architectural Acoustics | Environmental Acoustics | Noise Control for Industry | Scientific Counsel | Why work with OAA? | Contact Us


I am an elementary teacher. It seems that in my classroom I have to almost shout to be heard. Can anything be done to improve the situation?

Two things determine how good your environment is for communication:

  • The amount of noise that competes with your speech
    The noise that competes with speech comes generally from two sources: the people in the room and their activity, and the ventilation noise.
  • The amount of reverberation in the room
    The reverberation in the room depends on the surfaces in the classroom that can reflect sound.

Generally, when we design classrooms or correct their acoustics, we first start with the ventilation noise and do everything possible to reduce it. Frequently, that requires that we change the type of ventilation unit in the room. Some ventilators are much noisier than others.

To reduce the level of noise from people and other sources in the room, we need to treat the surfaces so that they absorb more sound. This sound absorption reduces the amount of vibration. Frequently, this is accomplished with a sound absorbent ceiling and sound absorbent panels or other finishes on the walls. However, it is important to recognize that placing too much sound absorption in a room or placing it in incorrect locations can work to the detriment of the ability to hear speech at great distances from the person talking.


How can you design a room for good communication?

First of all, we know the levels at which people normally speak. With this knowledge, we can calculate how much competing noise can be tolerated in the room. Based on this information, we can calculate a measure of communication adequacy which is known as the Articulation Index. his is a number that ranges from 0 to 1 and is a quantitative indicator of how well speech can be understood. An Articulation Index greater than 0.7 is acceptable for classroom communication.

An Articulation Index of 0.3 or less means that speech will not be understood. After the Articulation Index is calculated, based on the level of speech and the level of noise in the room, we make additional adjustments based on the reverberation time of the room. The two graphs below (?????) show the relationship between the Articulation Index, the speech-to-noise-ratio, and the percent of words understood. A second graph shows how the increase in reverberation time degrades the Articulation Index. Basically, in addition to an Articulation Index of .7 or better, we generally design a room so that the reverberation time is less than 500 ms.


If I can only make a small improvement in the acoustics in this room, will that have much of an effect on the ability of the children to hear me?

If the room is inadequate for communication to begin with, then small improvements in the Articulation Index can result in large improvements in the percentage of words understood correctly. If you look at the graph of the last question (???), you will note that a change in the Articulation Index from .3 to .4 can result in a change from 40 percent to 60 percent of the words understood correctly. A change from .3 to .5 represents an increase from 40 percent to 75 percent words understood correctly. This means that small changes in the acoustics can result in large improvements of speech intelligibility.


Someone has suggested that I could plant a row of trees to reduce the noise emitted from my factory to a nearby residence. How much of an improvement can I expect?

Understand that sound travels through air. The only way to reduce the amount of sound that reaches a listener is to create an obstruction between the listener and the source. Think first about decisuous trees in the winter. There are trucks and bare branches, through which it is relatively easy to see somebody who might be listening. Because there are no obstructions between the source of the sound and the listener, you cannot expect to achieve any meaningful sound reduction from these trees during the winter. Similarly, during the summer, when the trees are in leaf, the sound still has small, direct pathways through which it can travel. In addition, the leaves themsleves do not absorb much sound, except at very high frequencies. Therefore, we would not expect to achieve much sound absorption from deciduous tree barriers.

Pine trees are a little bit different. Their leaves are much denser, and therefore there is more of an obstruction in the sound transmission path. However, most pine trees do not grow close to the ground, therefore you can usually transmit sound to a listener along the open underside of the trees. For dense pines growing close to the ground, you might expect to achieve as much as 0.03 decibels of sound reduction for every foot of barrier. Thus, for a 100-foot-wide tree barrierk you would only expect to achieve 3 decibels of sound reduction. This is not an effective method of noise control.


Are there any regulations that I can rely on that limit sound emissions from other sources to my property?

In the State of New Jersey, there are limits that are set for sound from industrial or commercial properties to commercial and residential receiving properties. Although these might be somewhat detailed, in general, the New Jersey Regulations limit the sound at the residential receiving property to 65 dB(A) during the day and 50 dB(A) at night between the hours of 10 p.m. and 7 a.m. At commercial receiving properties, these are limited to 65 dB(A) at all times of the day.

Sound emitted to industrial land uses is not covered by the regulation.


How loud must a sound be for me to hear it?

This depends on your hearing ability and on the amount of competing or masking sound. For normal hearing people, the higher the ambient noise, the more intense a sound needs to be in order to be heard. The level of the sound that you want to hear must be greater than the level of the ambient sound in a credible band.


How intense a sound is needed for it to provide an adequate warning of danger or some other situation?

To understand the evaluation of warning signals, first consider the definition of the three terms: undetectable, detectable, and audible.

  • A sound which is undetectable elicits no response from the listener. It is not heard.
  • A sound which is detectable is one to which the listener will likely respond, "I think I hear something." Identification of the sound might not be possible, so an appropriate response to that signal is unlikely.
  • An audible sound is one which is clearly identifiable. As such, the listener will respond appropriately, by moving out of the way of a hazard or comforting a crying child, for example.

Several studies of signal audibility have reached similar conclusions: for a signal to be detectable to the "average" listener requires that the level in at least one-third octave band exceed the background sound pressure level in that band by 6 dB. This means that warning signals should be at least 15 dB above the background sound pressure levels in at least one-third octave band. The greater the number of bands in which the signal exceeds the ambient by the criterion 15 dB, the more audible will the signal be.